Reaching Beyond the Ivory Tower: A “How To” Manual *

Daniel Byman and Matthew Kroenig

Security Studies (forthcoming, June 2016)

*For helpful comments on earlier versios of this article, the authors would like to thank Michael C. Desch, Rebecca Friedman, Bruce Jentleson, Morgan Kaplan, Marc Lynch, Jeremy Shapiro, and participants in the Program on International Politics, Economics, and Security Speaker Series at the University of Chicago, participants in the Nuclear Studies Research Initiative Launch Conference, Austin, Texas, October 17-19, 2013, and members of a Midwest Political Science Association panel. Particular thanks to two anonymous reviewers and the editors of Security Studies for their helpful comments.

1 Joseph Nye, one of the rare top scholars with experience as a senior policymaker, lamented “the walls surrounding the ivory tower never seemed so high” – a view shared outside the academy and by many academics working on national security.1 Moreover, this problem may only be getting worse: a 2011 survey found that 85 percent of scholars believe the divide between scholars’ and policymakers’ worlds is growing. 2 Explanations range from the busyness of policymakers’ schedules, a disciplinary shift that emphasizes theory and methodology over policy relevance, and generally impenetrable academic prose. These and other explanations have merit, but such recommendations fail to recognize another fundamental issue: even those academic works that avoid these pitfalls rarely shape policy.3

Of course, much academic research is not designed to influence policy in the first place.

The primary purpose of academic research is not, nor should it be, to shape policy, but to expand the frontiers of human knowledge. For many academics, the creation of original research and gaining the esteem and recognition of their fellow scholars is gratification enough. For such scholars, this article may be of little use and there is no reason to read further. Many other

1 Joseph S. Nye Jr., “Scholars on the Sidelines,” Washington Post, April 13, 2009, http://www.washingtonpost.com/wp-dyn/content/article/2009/04/12/AR2009041202260_pf.html and Nicholas Kristof, “Professors, We Need You!,” New York Times, February 15, 2014, http://www.nytimes.com/2014/02/16/opinion/sunday/kristof-professors-we-need-you.html?_r=0 2 Paul Avey, Michael Desch, James D. Long, Daniel Maliniak, and Michael J. Tierney, “Inside the Ivory Tower IV,” Foreign Policy (January/ February 2012), 66. 3 For valuable takes on this issue, see among others, Alexander George, Bridging the Gap: Theory and Practice in Foreign Policy (Washington, D.C.: U.S. Institute of Peace, 1993); Alexander George, Presidential Decisionmaking in Foreign Policy: The Effective Use of Information and Advice (Boulder, CO: Westview Press, 1980); Bruce W. Jentleson, “The Need for Praxis: Bringing Policy Relevance Back In,” International Security 26, no. 4 (Cambridge, Massachusetts: Belfer Center, Spring 2002): 169-83; Joseph Kruzel, “More a Chasm Than a Gap, but Do Scholars Want to Bridge It?” Mershon International Studies Review 38, no. 1 (1994): 38; Joseph Lepgold and Miroslav Nincic, Beyond the Ivory Tower: International Relations and the Pursuit of Policy Relevance (New York: Columbia University Press, 2001); Ian Shapiro, The Flight from Reality in the Human Sciences (Princeton: Princeton University Press, 2005); Henry Farrell, “Why Do Policy Makers Hate International Relations Scholarship,” September 18, 2013, The Monkey Cage , September 18, 2013, http://themonkeycage.org/2013/09/18/why-do-policy- makers-hate-international-relations-scholarship/; Stephen M. Walt, “The Relationship between Theory and Policy in International Relations,” Annual Review of Political Science 8 (2005): 29-32; Philip Zelikow, “Foreign Policy Engineering: From Theory to Practice and Back Again,” International Security 18, no. 4 (Spring 1994): 143-71. An especially valuable recent work is Paul Avey and Michael Desch, “What Do Policymakers Want from Us? Results of a Survey of Current and Former Senior National Security Decision-makers,” International Studies Quarterly 58, no. 4 (December 2014).

2 scholars, however, are drawn to political science research in the hope that their ideas may influence important policy debates. According to the most recent TRIP survey, 60% of American

IR scholars have sought “to make their research more relevant to policy practitioners.” 4 Indeed, there is no reason why at least some subset of the international relations scholarship produced cannot both generate novel theoretical and empirical contributions and simultaneously inform foreign policy debates.

In order for this to happen regularly, however, scholars need to be cognizant of the types of scholarship that are useful to policymakers, the conditions under which outside ideas can matter, and steps academics can take to bring their work to the attention of policymakers, as well as the broader barriers that make policymakers ignore academic work in general. The questions of what , when, and how outside ideas receive a hearing are particularly important for those scholars who are attempting to be relevant: it is one thing to write only for academics and find that policymakers ignore one’s work; it is far worse to try to write for policymakers and still be ignored.

This article is a think piece. It tries to identify the conditions under which ideas of academics – and non-government knowledge in general – are more likely to be considered by and influence those in the policy world. To reach beyond the ivory tower, we argue that scholars should: design research that might produce actionable findings and recommendations, identify moments of ripeness in policy debates, and inject their ideas into the policy process. In particular, certain types of research, such as studies with relevant dependent variables and manipulable independent variables are most helpful to policymakers. To be relevant, scholarship must also be timely, appearing when the decision making process is ferment. The article also

4 Sue Peterson, Mike Tierney, Dan Maliniak, and Ryan Powers, “Teaching, Research, and International Policy (TRIP) 2014 Survey of International Relations Scholars,” available at https://trip.wm.edu/reports/2014/rp_2014/index.php.

3 argues that the conditions under which academic ideas are most likely to matter include discontinuous events or shocks, such as the 9/11 attacks; policy failures, such as the outbreak of an insurgency in Iraq; and unexpected decisions for which there is little initial government knowledge (such as the U.S. decision to intervene for humanitarian purposes in Somalia).

Finally, the article argues that there are concrete steps, including creating personal relationships with policymakers, positioning one’s work in the bureaucratic fray, and writing short, spin-off pieces for non-academic outlets with the policy implications clearly spelled out, that academics can take to disseminate their ideas into the policy process. Unfortunately, scholars who want to influence policy often do not design research appropriately or take advantage of these conditions and opportunities to advance their ideas.

This article draws on a range of empirical sources. These include an analysis of policymaker memoirs and writings to understand from where they get their ideas. 5 In addition, the authors of this paper conducted several interviews of individuals in 2013-2014 who serve or have served in the policy community, with an emphasis on those with an academic background – a “best case” for the influence of outside ideas in the policy world. In addition, this paper draws upon the relevant experiences of both authors in serving numerous government agencies and in attempting to conduct policy-relevant research. From these sources, this article advances several propositions about why certain ideas have a policy impact while others do not.

5 See Madeleine Albright, Madam Secretary: A Memoir (Hyperion, 2003); James Baker III, The Politics of Diplomacy (G.P. Putnam’s Sons, 1995); Joe Biden, Promises to Keep: On life and Politics, (Random House, 2008); George H. W. Bush and Brent Scowcroft, A World Transformed (New York: Vintage, 1998); George W. Bush, Decision Points (New York: Broadway Paperbacks, 2010); Richard Cheney, In My Time (New York: Threshold Editions, 2011); William J. Clinton, My Life (New York: Vintage, 2005); Robert Gates , From the Shadows: The Ultimate Insider’s Story of the Five Presidents And How They Won the Cold War (Simon & Schuster, 2007); My Life (New York: Vintage, 2005); Douglas Feith, War and Decision: Inside the Pentagon at the Dawn of the War on Terrorism (Harper Perennial, 2006); Chuck Hagel, America: Our Next Chapter: Tough Questions, Straight Answers , (Ecco, 2008); Robert Gates, Duty: Memoirs of a Secretary at War (New York: Knopf, 2014); Colin Powell, My American Journal (New York: Ballantine Books, 1995); Condoleezza Rice, No Higher Honor (New York: Crown Publishers, 2011); Donald Rumsfeld, Known and Unknown (Penguin, New York: 2011); George P. Shultz, Turmoil and Triumph: My Years as Secretary of State (New York: Charles Scribner’s Sons, 1993); and George Tenet, At the Center of the Storm: The CIA During America’s Time of Crisis (Harper Perennial, 2008).

4 The remainder of this article has seven parts. First, we define our key concept: policy relevance. Second, the article makes the case for policy relevance, arguing both for its value to the policymaker and contending that it improves the work of scholars. Third, we review existing arguments on why international relations scholarship is rarely policy relevant and propose several additional factors of our own. Fourth, this article argues that there are important exceptions to this bleak picture – and that these exceptions can become the norm if academics are willing to recognize them and change their work accordingly. We then advance a set of propositions or guidelines for those seeking policy relevance. The sixth section provides an example of rigorous scholarly research that became relevant to deliberations at the highest levels of the U.S. government. The final section offers concluding remarks.

I. Defining Policy Relevance

It is important to begin with a definition of policy relevance. We believe that much of the pessimism about the irrelevance of academic international relations scholarship is the result of unrealistic conceptions about what it means to be policy relevant. When academics say that they should have more of an influence on policy, they generally do not mean that they would like to weigh in on details of the policy process (who should draft the President’s directive on signals intelligence?), program management (which part of the State Department should train peacekeeping forces?), or policy implementation (what is the best rotation schedule for forces that will remain in Afghanistan until 2016?). Rather, they yearn to have a decisive influence on the formulation of the strategic options for the most important geopolitical issues facing the

5 country. Should we intervene in Syria? Can we live with a nuclear-armed Iran? What is the proper mix of engagement and containment for dealing with a rising China?

By this standard, however, the vast majority of national security policymakers are themselves not policy relevant. Nearly all officials in the national security bureaucracies are working on other, less salient, issues. Many are involved in what former Secretary of State

George Shultz referred to as “gardening” – maintaining organizations, checking in on friends, and ensuring the success of existing policies. 6 Even when officials are working directly on a hot- button subject, they might be excluded from the process because it is so sensitive that the senior- most officials want to keep deliberations close hold. 7 And, finally, even if one is part of the process on the most sensitive subjects, one is not generally welcome to take a step back and lay out a strategic vision for the country. As former Secretary of Defense Robert Gates wrote,

“People at lower levels had good ideas, but they had an impossible task of breaking through the bureaucracy, being heard and being taken seriously.” 8 Rather, one staffs one’s boss and hopes to have an influence by subtly shaping her thinking. She in turn might shape her boss, who shapes his boss, and so on, all the way to the top. Therefore, it is only the top national security officials in the nation who routinely have a decisive influence on the most important strategic decisions facing the country. Academics cannot expect to have a greater influence on policymaking than the vast majority of policymakers. Rather, a more realistic standard is needed.

We define policy-relevant academic scholarship as that which produces ideas that feature in the deliberations of senior government officials as they weigh policy decisions . To be policy relevant, an idea does not necessarily have to be the decisive factor in setting policy. Such a

6 Shultz, Turmoil and Triumph, 128. 7 Elliott Abrams, “Bombing the Syrian Reactor: The Untold Story,” Commentary, February 1, 2013, http://www.commentarymagazine.com/article/bombing-the-syrian-reactor-the-untold-story/ 8 Gates, Duty , 133.

6 standard sets too high a bar and would likely consign all academic scholarship and the vast majority of policymakers themselves to the fate of irrelevance. Rather, an academic idea can be relevant even if the final policy decision is contrary to scholar’s recommendation, as long as the idea was part of the conversation.

Ideas can infiltrate the process directly if a senior government official reads and is influenced by a piece of scholarship, but much more likely is that the academic argument or finding influences the process indirectly through variety of mechanisms: the scholar may write a policy piece or op-ed based on the scholarship that is read by officials, the scholar may brief policymakers on his or her academic findings, the scholarship may influence intelligence analysts or lower-level government officials who include its key findings or ideas in intelligence analysis, read-aheads, or talking points for more senior government officials, it can show up in the framework that policymakers use to conceptualize breaking events and in their interpretation of how a current policy is progressing, policymakers may be exposed to the idea years before as students or as outside analysts and bring the idea with them when they enter government service, or through a variety of other indirect channels discussed below.

On the other hand, if scholarship does not come to the attention of policymakers, or the idea comes to their attention, but is ignored because it is irrelevant or impractical, then that research cannot be considered policy relevant.

One can therefore think of international relations scholarship as falling on a spectrum from least to most policy relevant: (1) some scholarship does not reach the radar screen of policymakers at all; (2) some scholarship comes to their attention, but is dismissed; (3) some scholarship becomes a part of serious analysis and deliberations when senior officials weigh policy options; (4) and, finally, some scholarship has a decisive bearing on the final choice of

7 policy. We define policy-relevant scholarship as scholarship that falls into categories 3 and 4 and the advice provided in this article is intended to help scholars push their research from categories

1 and 2 and into the latter categories.

Others have defined policy-relevant scholarship as research that establishes “an explicit priority of addressing policy questions,” but we believe our definition to be superior because academic research can be relevant to policy even if that was not the explicit priority of the researcher and because much research that prioritizes policy relevance fails to achieve its mission.9

We should be clear that policy relevance is not the same as importance. Ideas can be important or influential for many other audiences even if they do not feature in the deliberations of senior policymakers. Moreover, scholarship on foundational issues, such as the nature of the international order or methodological developments, advances our knowledge in general and is valuable in and of itself. Policymakers are interested in a fairly narrow range of issue areas, and they may even often be wrongheaded in their focus. Nevertheless, our interest in this article is on policy relevance as defined here, not some broader conceptions of importance.

As American international relations scholars, we are most interested in ideas that influence U.S. foreign policy and, given the executive’s outsize role in setting foreign policy in the U.S. system, we focus mostly on how ideas penetrate decision making in the executive branch. We believe, however, that many, if not all, of our recommendations can be applied by scholars seeking to influence other policy audiences, including legislative branches, foreign governments, nongovernmental organizations, and international institutions.

9 Bruce W. Jentleson and Ely Ratner. “Bridging the Beltway–Ivory Tower Gap.” International Studies Review 13, no. 1 (March 2011): 6–11.

8 II. Why Bother?

The relevance of policy to the discipline of political science is both obvious and tricky.

Not all political science should be of immediate concern to policy leaders: every field has basic research, and it is often integral to the advancement of applied work. Moreover, as we will discuss below, there are many downside risks to guard against when pursuing policy relevance.

Nevertheless, we believe that there are a number of virtues to academics pursuing policy relevant research. Not least among them is that irrelevance risks the loss of resources. S enators

Tom Coburn and John McCain in 2013 changed the authorization for National Science

Foundation funding for political science, requiring it to be relevant to national security and U.S. economic interests.10 Even if dollars continue to flow, however, many scholars, ourselves included, believe that our ideas and knowledge offer insights that could improve the craft of foreign policy, better ensure our security, manage and maintain robust institutions, and otherwise make our countries, and our world, better places. Alexander George points out that academics can help policymakers identify strategies and determine their requirements, assess the strengths and weaknesses of different strategic logics, and develop nuanced approaches to particular actors in the system. 11 Indeed, many scholarly journal articles conclude with an explicit (although often brief and imprecise) discussion of the real-world implications of the research in an attempt to do just that.

Many scholars study international relations because of their personal interest in international affairs and the great decisions being made about them. As the survey cited in the introduction suggests, a majority of IR scholars want to help shape these decisions for their own

10 For a discussion, see Gregory Koger, “Political Science, NSF Funding, and the National Interest,” The Monkey Cage, March 22, 2013, http://themonkeycage.org/2013/03/22/political-science-nsf-funding-and-the-national- interest/ 11 George, Bridging the Gap, 117-31.

9 satisfaction. Perhaps more important, universities are also crucial institutions in society, and there are calls for universities to speak to daily issues and practical needs of their country and the world in general. Such calls for “relevance” often go too far and diminish the value of traditional research, but having more policy-relevant work would help universities maintain their central role in society. 12

The work of scholars offers unique benefits to policymakers. Although think tanks, lobbyists, and government bureaucracies themselves offer information to policymakers, scholars can take advantage of our time to contemplate, our rigorous training in weighing ideas, and, especially, our relative bureaucratic independence and neutrality. Scholars can more easily voice contrarian ideas (“more nuclear proliferation may be better” or “humanitarian intervention inadvertently leads to mass killing”) that are difficult for those in the system to put forward. 13 As scholars have withdrawn from the policy world, think tanks play a greater role in transmitting ideas. David Featherman, the former president of the Social Science Research Council, noted bitterly the irony that academics “in their quest for professional integrity and scientific objectivity” have yielded the field “to the private, often partisan think tanks.” 14

Less obviously, speaking to policymakers has value for scholarship and can result in a benign circle: by engaging the policy world we better understand what we study and do work

12 See, for example, Nicolas Lemann, “The Soul of the Research University,” The Chronicle of Higher Education, April 28, 2014 http://chronicle.com/article/The-Soul-of-the-Research/146155/. For a criticism, see Terry Eagleton, “The Slow Death of the Research University,” Chronicle of Higher Education, April 6, 2015, http://chronicle.com/article/The-Slow-Death-of-the/228991/ 13 See, for example, Kenneth Waltz, “The Spread of Nuclear Weapons: More May Better,” Adelphi Papers , no. 171 (London: International Institute for Strategic Studies, 1981) and Alan Kuperman, “The Moral Hazard of Humanitarian Intervention: Lessons from the Balkans,” International Studies Quarterly 52, no. 1 (March 2008): 49- 80. For the difficulty of dissent among the inner circle, see George, Presidential Decisionmaking in Foreign Policy, 81-108. 14 Cited in Bruce W. Jentleson and Ely Ratner, “Bridging the Beltway-Ivory Tower Gap,” International Studies Review 14, no. 1(2011): 7-8. For more on the role of think tanks, see James McGann and R. Kent Weaver, eds, Think Tanks and Civil Societies: Catalysts for Ideas and Action (New Brunswick, NJ: Transaction Publishers,) 2002.

10 that is of higher quality in general and increased relevancy for policymakers.15 As Joseph

Lepgold pointed out, realism was first a set of maxims for statecraft and only later became an academic theory as practitioners put their ideas down more formally.16 In addition, scholars can formulate better research questions with greater exposure to the real-world puzzles with which policymakers must grapple. The questions that emerge from vexing foreign policy challenges often make for as good, if not better, scholarly research agendas as those that emerge from gaps in the academic literature. Furthermore, Bruce Jentleson and Ely Ratner contend that one’s own experience can be used to absorb and observe and, in so doing, be a form of limited theory testing. 17 Putting theorizing aside, being closer to policy makes it more likely that scholars will understand the data on which they rely and encourages intellectual pluralism. As scholars embrace policy concerns, the policymakers’ preoccupations, concerns, foibles, and worldviews all become clearer – and thus enhance our research on international relations. For example, as

David Newsom points out, scholars often rely on documents for their data, but memoranda of conversation, reports to the President, and other key documents are often written after decisions are made and are part of a deliberate paper trail to spin a policy against a media leak or

Congressional inquiry. Data from interviews similarly try to present one’s own actions in a positive light. 18 Policy experience can help scholars understand and guard against these biases.

III. The Irrelevance of International Relations Scholarship

15 See R. Charli Carpenter, “You Talk of Terrible Things So Matter-of-Factly in This Language of Science”: Constructing Human Rights in the Academy,” Perspectives on Politics 10, no. 2 (June 2012) for a persuasive argument on how participation can improve understanding of the issues a scholar studies. 16 Joseph Lepgold, “Is Anyone Listening? International Relations Theory and the Problem of Policy Relevance,” Political Science Quarterly 113, no. 1 (1998): 45. 17 Jentleson and Ratner, “Bridging the Beltway-Ivory Tower Gap,” 8. 18 David D. Newsom, “Foreign Policy and Academia,” Foreign Policy, no. 101 (Winter, 1995-1996): 56.

11 Despite the benefits for both sides, most international relations scholarship remains policy irrelevant. Paging through the top journals in international relations, it is easy to question the policy relevance of much of the content. Indeed, a survey of the works on the relationship between the policy and academic worlds indicates widespread agreement on academia’s policy irrelevance. James Kurth pointed out in 1998 that international relations scholars are “Rightly ignored by practicing foreign policy officials.” 19 A decade later, Nye would mourn the growing gap and the paucity of scholars in senior ranks of government. 20 One former official we interviewed who had scholarly credentials herself summed up what seems to be the current wisdom, lamenting that “the discipline of IR is divorced from the real world.” 21 There are of course exceptions, but these are just that. The leading scholars cited in a survey conducted by

Paul Avey and Michael Desch included eminent figures like Henry Kissinger, Francis

Fukuyama, and Zbigniew Brzezinski. These figures were often important in large part due to their own prominent policy role outside of the academy and have written less for scholarly and more for public audiences. 22 That study found, “A small circle of scholars makes their views known in op-eds or blog posts, or by taking sabbaticals inside the Beltway, but the views of most academics remains unheard in Washington.” 23

Self-Inflicted Sources of Irrelevance

19 James Kurth, “Inside the Cave: the Banality of IR Studies, The National Interest 53 (1998): 29-40; Nye, “Scholars on the Sidelines.” 20 Joseph S. Nye Jr., “Scholars on the Sidelines,” Washington Post, April 13, 2009 21 Interview with official who had served in the Obama administration, Washington D.C., December 15, 2013. 22 For the list of influential names, see Avey, et al., “The Ivory Tower Survey.” Joseph Nye would be an exception to this point. 23 Avey et. al, “The Ivory Tower Survey.” American scholars estimate they put only 6 percent of time into non- academic consulting.

12 A number of scholars have sought to explain why academic works are for the most part ignored and, in this section, we compile some of them and add some of our own views. Sources of irrelevance include: impenetrable prose, an emphasis on methods over content, a lack of awareness of empirics and the policy process, unresolved debates in the field, work that appears well after a decision is made, the discipline’s incentive structure that discourages policy-relevant work, and academics’ high level of generality

Impenetrable prose. Academic writing is often bad. 24 Part of this problem might be sloppiness, but the deeper problem is the regular use of jargon – which, as Nye notes, “is unintelligible to the policymakers.”25 Specialized language is a shortcut for specialists and thus has value – if we have to define methodological terms that are obscure, like “endogeneity” (let us put it this way – our spell checker doesn’t recognize it), or specialized ones, like “anarchy,” which poets and New York Times readers define differently than do scholars, then many articles would be weighed down by tedious and repetitive terminology reviews. Yet by choosing to write with jargon we are also choosing to exclude non-specialized audiences. In addition, inventing new terms and other forms of jargon often cover up weak logic with a mask of sophistication.

Methods Bias. A related problem is the trend in political science to emphasize methods over content – a leading source of irrelevance according to Avey and Desch. Indeed, one participant in their survey summed up formal models as “a complete waste.” 26 Concerns over endogeneity (see the jargon point, above) lead to an emphasis on pristine research designs that avoid the muddiness of studying actual policy, where causality can run in many directions. The use of large datasets often requires simplified metrics (e.g. coding all military interventions into

24 Patricia Nelson Limerick, “Dancing with Professors: The Trouble with Academic Prose,” New York Times Book Review, October 31, 1993, http://www.cs.tufts.edu/~nr/cs257/archive/patricia-limerick/dancing.htm 25 Nye, “Scholars on the Sidelines.” 26 Avey and Desch, “What Do Policymakers Want from Us?”

13 simple binary categories like successes versus failures – a difficult task when we still debate this question on many wars), and the presentation of these materials in statistical form often strikes policymakers as confusing, unfaithful to a complex reality, or both. 27 As a result, it is hard for policymakers to see a resemblance between the real problems with which they grapple and the abstract subjects treated in the scholarly literature. This simplification often occurs to the point of absurdity or amusement from a policy point of view – Secretary of State Dean Acheson, on learning that his role in supporting the U.S. intervention in Korea was treated as a dependent variable, wryly took exception and declared that he thought of himself as an independent variable. 28 To be sure, advanced methods and quantitative analysis can be essential to analyzing some policy issues. For example, research on whether numbers of nuclear weapons matter (a case study to which we will return below) calls out for a quantitative assessment. Still, according to the most recent TRIP survey, only 9% of American IR scholars employ “policy analysis” as a

“primary methodology,” even though 88% said such research was either “somewhat useful” or

“very useful” to policy practitioners. 29 Since the vast majority of scholars are choosing a methodological approach for reasons other than its usefulness to policymakers, then it should come as no surprise that much academic research does not address policymaker concerns.

Uninformed. International relations scholarship also can be remarkably uninformed on both empirics and the policy process. The emphasis on theory and methods often comes at a price in terms of detailed case-based and historical knowledge and the formulation of policy.

Moreover, scholars often intentionally overlook substantive empirical knowledge in a quest for theoretical elegance and parsimony. Lawrence Mead argued that, “[f]or many, politics or

27 Ibid 28 George, Bridging the Gap , 7. 29 Peterson, Tierney, Maliniak, and Powers, “TRIP 2014 Survey of International Relations Scholars.”

14 government is something to model, not important in itself.” He further laments that, “To them, the subject is the literature.” 30

Irresolution. One virtue of scholarship is that it questions assumptions. The costs of democracy in fighting terrorism, the benefits of military victory over negotiation for lasting peace, and the overblown reaction to terrorism are only a few examples of important scholarly arguments that flout what is politically correct and widely believed. 31 This virtue can become a vice in terms of policy influence. When an academic debate is not resolved, it allows proponents and opponents of a policy to claim scholarship supports their approach or to declare it irrelevant. 32

Untimely. For policymakers, being timely is as important as being accurate –the right answer too early is ignored; too late, it is of no use. As Nye contends, if a product is late in the policy world it is “a total failure.” 33 As part of our goal is to teach and explain, accuracy should be at the core of our discipline. Yet for work to matter for policy, it must arrive in time for it to be considered as policy is formulated. Colin Powell notes that he could not wait for enough facts to be 100 percent sure he was right because by the time the facts arrived it would be too late to act. 34 In addition, the beginning and end of a policy formulation period is nebulous – policies are constantly being reevaluated and the same goals are often implemented in different ways – but once the President and his most senior advisors weigh in and programs and resources are

30 Lawrence M. Mead, “Scholasticism in Political Science,” Perspectives on Politics 8, no. 2 (June 2010): 454. 31 F. Gregory Gause, “Can Democracy Stop Terrorism?” Foreign Affairs (September/October 2005), http://www.foreignaffairs.com/articles/61021/f-gregory-gause-iii/can-democracy-stop-terrorism; Monica Duffy Toft, “Ending Civil Wars: A Case for Rebel Victory?” International Security 34, no. 4 (Spring 2010): 7-36; John Mueller and Mark Stewart, “The Terrorism Delusion: America’s Overwrought Response to September 11,” International Security 37, no. 1 (Summer 2012). 32 Lepgold, “Is Anyone Listening?” 46; Jeremy Shapiro and Miriam R. Estrin, “How the U.S. Saw Syria’s War,” December 5, 2013, http://foreignpolicy.com/2013/12/05/how-the-u-s-saw-syrias-war/ 33 Nye, 117, in Krasner, Nye, Stein, and Keohane, “Autobiographical reflections on bridging the policy-academy divide.” 34 Powell, My American Journey, 393.

15 committed, it is far harder to effect change. 35 Too often, however, academics wait for enough data to be amassed before joining the debate. Others may join in after a presidential announcement (say whether or not to intervene in Syria) even though by then the bureaucracy has its instructions and it is far harder to shape outcomes.

For academics, the preferences and publication schedules of top-ranked journals are in opposition to policy timeliness. 36 At a minimum, publication in say the American Political

Science Review takes nine months from drafting to publication – peer review and the resulting revisions add considerable time for even the fastest writers. And nine months is a heroic assumption. From our experience, to make an article strong enough to have a chance of being accepted to a top journal, a year of collecting data, doing analysis, drafting, presenting at conferences, sharing with colleagues, and revising is likely. Then, after submitting, three months is necessary – though six or even nine months are more common trajectories. An encouraging

“revise and resubmit” adds additional time. Should it be accepted, a likely backlog in most top journals further delays things. In the end, two years seems a minimum, with three to four years being more common. As Bennett and Ikenberry point out, if a scholar had decided to write on the politics of Congressional authorization of U.S. military intervention in Iraq in 2002, even under the best circumstances the war would have begun before the article was published. 37 In the more likely case, the associated insurgency and civil war would be raging, making the argument seen overtaken by events.

Incentive Structure. Advancement in the field, which depends so much on publication in top journals, compounds the above set of problems. Simply put, journals that matter for

35 Bush and Scowcroft, A World Transformed, 211. 36 Lepgold, “Is Anyone Listening?” 45. 37 Andrew Bennett and G. John Ikenberry, “The Review’s Evolving Relevance for U.S. Foreign Policy, 1906-2006,” American Political Science Review 100, no. 4 (November 2006): 653.

16 promotion publish few articles that have immediate policy relevance. Daniel Maliniak, Amy

Oakes, Susan Peterson, and Michael J. Tierney find that few top journals offer policy advice

(though International Security and Security Studies are exceptions) – and that most of their policy relevance comes after a major event, not before it. In addition, strategic regions are often neglected. 38 A study of the American Political Science Review found that since the 1960s, “the

Review has published fewer and fewer articles aspiring to near-term or direct policy relevance” – and that other journals are not picking up the slack. 39 In their private reflections, many scholars recognize this problem. One survey found that almost half of political scientists rarely found anything of interest in their leading field journal, a higher figure than that of any other field. 40

As the field does not value real-world relevance, it promotes accordingly, creating a malign circle. 41 As Nye contends, “Scholars are paying less attention to questions about how their work relates to the policy world, and in many departments a focus on policy can hurt one’s career.” 42 Many programs, particularly top ones, emphasize “more about the discipline than the world,” focusing on theoretical debates, game theory, and statistics over knowledge about policy areas or particular regions. 43 An article in a more policy relevant journal like Foreign Affairs, let alone an opinion piece, is usually ignored or even denigrated in the tenure process. When hiring, many search committees focus on theoretical originality and methodological sophistication, not policy significance. Tenure codifies this, and the circle continues. 44

Different Approaches to the Problem. Academics and policymakers work on the same subject matter, but they approach these problems in different ways and this limits academics’

38 Maliniak, et a l., “International Relations in the US Academy,” 457 and 459. 39 Bennett and Ikenberry, “The Review’ s Evolving Relevance for U.S. Foreign Policy,” 652. 40 Mead, “Scholasticism in Political Science,” 461. 41 Walt, “The Relationship between Theory and Policy in International Relations,” 38-39. 42 Nye, “Scholars on the Sidelines.” 43 Jentleson and Ratner, “Bridging the Beltway-Ivory Tower Gap,” 7. 44 Ibid; Jentleson, “The Need for Praxis,” 178.

17 ability to speak to the concerns of policymakers. For example, academics often study the relationship between a single independent variable and a single dependent variable (e.g., does nuclear proliferation increase or decrease international stability?), whereas policymakers must consider how developments could affect a broad array of U.S. interests (e.g., how does nuclear proliferation effect: regional and international stability, the risk of nuclear war, the foreign policy of the proliferant state, US freedom of action, U.S. credibility, the security of allies, oil markets, the nonproliferation regime, etc.?) Relatedly, academics are interested in generalizable relationships produced by the average case, whereas policymakers care about the specific case.

Recent scholarship has shown that nuclear powers are no less conflictual than nonnuclear powers on average, but U.S. policymakers want to know how Iran will behave with nuclear weapons.45

As Kruzel contends, “A bureaucrat wants specific advice, not general theory.” 46 After all, there

are outliers and exceptions to any academic finding and given that policymakers focus their time

and attention on the most threatening or intractable cases, chances are they are outliers – or at

least perceived as such. In addition, policymakers and academics think differently about risk.

Academics attempt to explain the world, but policymakers are responsible for protecting the

national security of the nation. It is possible, therefore, that academic studies can demonstrate

that the objective probability of nuclear war is low and that policymakers should still devote

much attention to the problem. 47 Furthermore, scholars tend to conduct either/or analysis

whereas policymakers want both/and analysis. Academics ask questions like: What is the more

important cause of nuclear proliferation, supply-side factors such as sensitive nuclear technology

45 Mark S. Bell and Nicholas L. Miller, “Questioning the Effect of Nuclear Weapons on Conflict,” Journal of Conflict Resolution 59, no. 1 (2015): 74-92. 46 Kruzel, “More a Chasm Than a Gap,” 179. Lepgold recognizes this point, pointing out that policymakers ask whether the cases on which a model is built are similar to those on their desks. Lepgold, “Is Anyone Listening?” 55. 47 Colin Kahl, “Policy Pessimism vs. Proliferation Optimism: The Case of Iran,” Paper Prepared for the Nuclear Studies Research Initiative Launch Conference, Austin, Texas, October 17-19, 2013.

18 transfers or demand-side factors like security environment? 48 If policymakers really want to prevent the spread of nuclear weapons and there is some reason to believe that efforts to deny technology and efforts to ameliorate security environments will both help, then they may do both, resources permitting. 49

IV. The Exceptional Exceptions: Opportunities for Relevance

All is not lost. Scholars’ irrelevance is not decreed, and policymakers at times will listen to outside ideas. Economics and medical research offer many instances where theories about monetary and fiscal policies, immunization effectiveness, and other issues go from academic journals to the policy world, albeit in an imperfect and fitful way. 50 Some top journals have published pieces of immediate and clear policy relevance (though it is too much to say that this is a trend). Further, a small but significant number of scholars do write for popular journals as well as scholarly publications, minimize jargon, and otherwise try to make their work more relevant.

Moreover, the Avey and Desch survey found that policymakers do want help in making sense of the world: theorizing, in another word. 51 The Council on Foreign Relations has long promoted the interaction of academics and the policy community by funding an exchange program.

48 Scott Sagan, “The Causes of Nuclear Weapons Proliferation,” Annual Review of Political Science 14 (June 2011): 225-44. 49 Scholars, of course, are not the only ones to blame. The policy side also is at fault. Policymakers often ignore valuable academic ideas and assessments because they: tend to devalue written analysis, lack adequate methodological training, are too busy to delve into academic studies, and because they rely on other sources of information, such as intelligence analysis. Also, although it is tempting to bemoan the irrelevance of international relations scholarship, it is vital to recognize that the problem is often worse for other subfields, academic disciplines, and in other countries. 50 Krasner, 112 in Krasner, Nye Jr., Stein, and Keohane, “Autobiographical reflections on bridging the policy- academy divide”; George W. Bush, Decision Points , 117-18. 51 Avey and Desch, “What Do Policymakers Want from Us?”

19 Promising efforts such as the Tobin Project, the Bridging the Gap Project, the Scholars’ Strategy

Network, the International Policy Summer Institute, “The Monkey Cage” and the work of Avey and Desch suggest that many scholars, both newcomers and established voices, recognize the need for policy relevant work and are trying to make it so. Prominent political scientist James

Fearon contends that political scientists are more cited in The New York Times than in the past, suggesting that critiques about the lack of engagement may be overstated. 52

There is no doubt that ideas from the outside do shape policy. The Heritage Foundation produced a massive report, Mandate for Leadership, at the outset of President Reagan’s term in office: of its 2,000 recommendations, roughly 60 percent were implemented. 53 George W.

Bush’s second term “freedom agenda” (which for several years shaped U.S. policy in the Muslim world) was influenced by the thoughts of Yale historian John Lewis Gaddis and the writings of

Soviet refusnik Natan Sharansky as well as Middle East scholars Fouad Ajami and Bernard

Lewis. In addition, the work of Arab intellectuals who prepared a report for the United Nations also shaped administration thinking. 54 The Iraqi surge and associated strategy shift was informed by a range of academic writings on counterinsurgency as well as General David Petraeus’ own work on the subject while studying at Princeton.55 Bennett and Ikenberry point to the Democratic

Peace Theory as an illustration of the “food chain” of how theories become policies. The scholarly article that pointed out that democracies do not fight one another was first introduced by Doyle in the American Political Science Review in 1986 found its way into Foreign Affairs

52 James Fearon, “Data on the relevance of political scientists to the NYT,” The Monkey Cage , February 23, 2014, http://www.washingtonpost.com/blogs/monkey-cage/wp/2014/02/23/data-on-the-relevance-of-political-scientists-to- the-nyt/ 53 Peter W. Singer, “Factories to Call Our Own,” Washingtonian, August 13, 2010, http://www.washingtonian.com/articles/people/factories-to-call-our-own/ 54 Peter Baker, Days of Fire: Bush and Cheney in the White House (2013); Rice, No Greater Honor, 326-27 . Baker reports that in the days leading up to his second inaugural address that announced the ambitious freedom agenda, Bush was heavily influenced and met with Sharansky about his book, The Case for Democracy: The Power of Freedom to Overcome Tyranny & Terror (Public Affairs, 2006). 55 Kaplan, The Insurgents.

20 and Foreign Policy and, eventually, into the rhetoric of Presidents H. W. Bush and Clinton, formally becoming part of the 1996 National Security Strategy. 56 President George W. Bush would similarly endorse this view as part of his freedom agenda. To be clear, in none of these cases was the outside idea the only, or even the primary, driver of the policy. Avey and Desch find the Democratic Peace Theory was known by 56 percent of policymakers surveyed (they consider this a small percentage, but it seems considerable to us).57 The above ideas mixed with the policy process at different stages and in different ways – but the specific ideas did influence the debate. 58

With these exceptions in mind, the question is not so much do ideas influence policy but rather what , when, and how. Below we advance propositions about the design, timing, and promotion of social science research that we believe affect their probability of policy influence.

Propositions on Design of Policy-Relevant Research. In order for scholarship to affect policy, one must begin by designing scholarship that can shed light on problems of interest to foreign policymakers and on what they can do to shape them. It is vital to have relevant dependent and independent variables, and qualitative work is more likely to have an impact.

P1. Relevant Dependent Variables. Scholarship that focuses on policy-relevant subjects

(or in social science terms, dependent variables) is more likely to influence policy. Scholars often

56 Bennett and Ikenberry, “The Review’ s Evolving Relevance for U.S. Foreign Policy,” 656. 57 Avey and Desch, “2011 Policymaker Survey,” 9. 58 Different administrations, of course, have different types of openings for outside ideas. One former official pointed out that President Clinton would famously scrawl on the margins of opinion pieces and demand answers from his aides. President George W. Bush was an avid reader and did “deep dives” into his intelligence, though often via briefing rather than writings. On a lower level, some officials such as the late Richard Holbrooke read extensively; others believe they already know all. So some of the policymakers can be engaged some of the time on some of the issues. Interview with former Clinton and Obama administration official, January 14, 2014, Washington, D.C.; Gates, Duty, 94. Secretary of Defense Hagel regularly reads history and from there extracts ideas. See Hagel, America, 66.

21 think that by working on important issues, such as war and peace, international stability, or grand strategy, they are being policy relevant. 59 But U.S. officials rarely, if ever, have the time or inclination to sit back and formulate “grand strategy” or to devise policies with the goal of contributing to international peace and stability. 60 Rather, there are many government officials who spend countless hours working on nuclear nonproliferation, arms control, counterterrorism, cybersecurity, Europe, the Middle East, Asia, and many other functional and regional issues.

This means that so called middle-level theory on these regional and functional subjects is of immediate interests to policymakers and academic ideas in these domains often feed into the policy process.61

There are no U.S. officials assigned with responsibility for peace, international stability, or grand strategy. Even the offices assigned with doing strategy, such as Policy Planning at State and the Strategy office at the Pentagon assign their staff to work on, and be more strategic in their approach to, the functional and regional issues on which policymakers spend most of their time.

As we will discuss below, occasionally windows open when more sweeping frameworks are needed to make sense of new eras or events (and the present may be one of those times) but, for the most part, scholarship on specific regional or functional problems will have more of an audience in the corridors of power.

P2. Relevant Independent Variables. This emphasis applies to independent variables as well. George laments studies that “do not include variables over which policymakers have some

59 The more abstract the idea, the farther it needs to travel to be relevant. Lepgold divides intellectual activities into four groups, ranging from pure theory to day-to-day efforts reflected in opinion pieces and Foreign Affairs. Over time he believed basic theory can make its way into applied theory and work focused on particular cases, which in turn shows up in more daily work. Lepgold, “Is Anyone Listening?” 60 Richard K. Betts, “Is Strategy an Illusion?” International Security 25, no. 2 (Fall 2000): 5-50. 61 See George, Bridging the Gap, for an argument along these lines.

22 leverage.” 62 The fundamental objective of scholars and policymakers is different. Scholars seek to explain the world. This means identifying important cause and effect relationships between independent and dependent variables. The goal of U.S. officials is to manipulate the tools at their disposal to advance U.S. interests. Many international relations scholars focus on how the international distribution of power, culture, levels of socio-economic development, and other systematic and structural factors affect international outcomes. Unfortunately, these are factors that U.S. foreign policymakers cannot influence (at least not easily) even if they tried – rendering their findings essentially useless in the policymakers’ toolbox. In contrast, other studies do focus on tools that policymakers can manipulate (force posture, mobilizations, threats, sanctions, military strikes, negotiations, etc.).

This is not to say that academics should ignore structural factors. Scholars must identify the most important cause-effect relationships and if an outcome is driven by structural factors, academics have a responsibility to accurately report their findings. Moreover, these findings can be helpful in instructing policymakers and intelligence analysts on the limits of what is possible.

At the same time, however, scholars should understand that there findings will be most relevant to the degree that they identify the factors that policymakers can manipulate to push outcomes in one direction or another.

P3. Method Matters . Scholarship that relies primarily on qualitative research and argumentation will be more likely to have an impact. This hypothesis borders on the obvious.

After all, it is difficult to reach an intended audience if that audience cannot understand the language in which one is attempting to communicate. If social science standards demand that a question be addressed through quantitative analysis, once can still use innumeracy to one’s advantage to convey expertise. Large-N scholarship can have an impact as long as the finding

62 George, Bridging the Gap, 8.

23 can be translated into plain English and illustrated with examples. After all, the bottom-line of most quantitative studies is fairly easy to digest if explained clearly; “X is associated with A” is not that difficult to understand. Another route is to write an oped or policy article based on the statistical finding (see the section on promoting academic research below).

Propositions on Timing of Policy-Relevant Research. Once a scholar has designed and conducted academic research, the central ideas, arguments, and findings the project generates will be more likely to find a receptive audience in the corridors of power at certain times rather than others. In general, ideas advanced early in a debate have more impact. Scholars can also seize on unexpected contingencies, discontinuous events, and policy failures to make it more likely their work will appear at the right time to matter.

P4. The Earlier the Better. Sometimes even good, policy relevant ideas are ignored. As

Samuel Lewis contends, “even after you build a bridge, there’s no guarantee anyone will use it.” 63 Too often academics engage the policy world after initial decisions are made. They may do so in jargon-free prose that focuses on variables that policymakers can manipulate, but “once policy is set, it is hard to change” as one former official remarked.64 When NATO enlargement was still being debated, outside work received serious attention: once National Security Advisor

Anthony Lake and President Clinton signed off on expansion, the outside work was then ignored. 65 Similarly, almost all academics opposed the U.S. invasion of Iraq, but serious work on the subject was scarce. This was not due to lack of time or warning: unlike most wars, serious thinking on this conflict began well over a year before the shooting started (and, one could argue, five years before). John Mearsheimer and Steve Walt’s work in Foreign Policy was one attempt

63 Samuel Lewis, “Introduction,” to George, Bridging the Gap, 10. 64 Interview with former Clinton and Obama administration official, January 14, 2014, Washington, D.C. 65 Ibid.

24 to engage this debate, but even this short piece appeared in the January/February 2003 issue of the magazine – many months after the President had decided to go to war. 66 In contrast, Kenneth

Pollack published a tome titled The Threatening Storm in September 2002 – offering a detailed engagement of issues relating to deterrence and rogue state behavior as well as drawing on his knowledge of Iraq. Such timely release enabled his work to have far more influence on the policy world. It was during these critical months that many policymakers formed their opinions,

Congress debated authorization, presidential policy became more set, and so on.

To be sure, as we argued above, time to adequately study an issue in depth is one of the greatest strengths of the academy and we are in no way arguing that academics should produce rushed or shoddy scholarship in order to meet the demands of policymakers. Rather, academics have a lifetime worth of scholarly findings and experience upon which they can draw in order to quickly inject important insights into the policy debate based on their reservoir of knowledge.

In addition, a case can be made for getting to policymakers before they are policymakers.

Government officials may have little time to think while in government, but when they are outside government ranks – perhaps at the academy or a think tank – they may read academic studies that contribute to conceptual thinking that they then employ in government. Over half of the Assistant Secretaries in the State Department came out of think tanks. 67 Nye recalls that his own thinking on China at Harvard led him to implement these ideas when he moved to the

Department of Defense: “intellectual capital and theoretical ideas produced and stored up in a university setting were useful in making policy changes in Washington.” 68

66 See John Mearsheimer and Steven Walt, “An Unnecessary War,” Foreign Policy (January/February 2003), http://mearsheimer.uchicago.edu/pdfs/A0032.pdf. 67 Singer, “Factories to Call Our Own.” 68 See Nye, 119 in Krasner, Nye Jr., Stein, and Keohane, “Autobiographical reflections on bridging the policy- academy divide.”

25 If policymakers’ mindset upon entering senior positions is shaped by their educations and experiences, then there is an ineffable, but important, effect of scholarship in shaping the policymakers years or decades earlier in their life – a variant of Keynes’ contention that

“practical men, who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist.” Hillary Clinton recalls that her views in college changed because: “My ideas, new and old, were tested daily by political science professors who pushed me to expand my understanding of the world and examine my own preconceptions just when current events provided more than enough material.” 69 Secretaries of State Albright and

Rice both attest to their academic training affecting how they saw the world when they became senior policymakers. 70 It would be hard to draw a causal line between a theory or argument learned in a class in 1975 and a policy implemented in 2005, and the information and experiences in the intervening decades may have shaped the lessons the individual learned and tried to implement while in office. Bennett and Ikenberry contend that basic political science theory is akin to basic research in medicine: it finds its way into more policy-driven scholarship, which in turn influences articles in Foreign Affairs or other outlines that embrace policy directly.

These, in turn, find their way into government documents. 71

This view, while comforting, may be wrong. One of the few studies of this issue, the

Avey and Desch survey, challenged the “trickle down” theory, finding that most policymakers cite professional experience, not education, in shaping their worldviews. 72 Policymakers may not fully recognize how their educations shape their worldviews, but with an absence of supporting

69 Rodham Clinton, Living History , 31. 70 Albright, Madam Secretary; Rice, No Greater Honor. 71 Bennett and Ikenberry, “The Review’ s Evolving Relevance for U.S. Foreign Policy,” 652. 72 Avey and Desch, “What Do Policymakers Want from Us?”

26 evidence academics should also not overstate how classroom teachings might affect real-world results.

P5. Seizing on Unexpected Contingencies. Governments are often unable to prepare for unexpected developments. At times, this failure to plan or anticipate may be political. For example, part of the reason for the dismal “Phase IV” planning for the Iraq war was a fear that by planning for the unanticipated (such as an insurgency), critics of the war would believe that this was in fact a likely event – and thus give them ammunition in their fight against going to war in the first place. During the first Gulf War, the United States did not plan for what to do after

Saddam’s forces were pushed out of Kuwait in part because of fears that this issue would be divisive among allies and that any planning would leak and undermine the coalition. Today, what would happen if it leaked to the press that Pentagon officials were developing a strategy to deter and contain a nuclear Iran even though Obama said that a nuclear Iran is “unacceptable?” While some might assume that the government should prepare for every contingency, the fact is that there are certain lines of inquiry that are politically unacceptable. And even when contingency planning is acceptable, scarce resources often are devoted to the present, not the possible.

Academics face no such constraints. Outside analysis on these issues may be wasted (the

North Korean regime, say, may never fall) but should the unexpected occur, often the outside analysis will be the only assessment initially available.

P6. Seizing Discontinuous Events. Once policy is set, a host of political and bureaucratic interests perpetuate it. The government itself often develops tremendous expertise on the perceived danger and the implementation of programs designed to combat it. Outside expertise may pale by comparison or, at the very least, has stiff competition against a policy that enjoys considerable momentum. At times, however, sudden events – the collapse of the Soviet Union,

27 the Arab spring, the 9/11 attacks, and so on – challenge current government assessments and policies and make existing in-house expertise seem less relevant and outside judgments more valuable. “What does the Rouhani election mean?” one former official asked in this context. 73 As one official contended, after the fall of the Berlin Wall, officials were caught flat-footed and

“looked to the academy for ideas” on what to do next. 74

It could be argued that we are currently at such a juncture. The wars in Iraq and

Afghanistan are winding down, new wars have erupted in Ukraine and against the Islamic State, and we have announced an intention to “rebalance” to Asia, yet many analysts point out that we have not yet fully departed from the Middle East and that the commitment to East Asia remains more of an aspiration than a reality. 75 In this context, what are the important events and foreign policy decisions that will shape the future of American foreign policy?

Discontinuous events may also occur due to surprising U.S. policy decisions. Often the

U.S. government acts without a deliberate policy process, such as the humanitarian intervention in Somalia. In such circumstances, new policies have not yet jelled (see “Earlier the Better”) and outsiders with insight into the unexpected challenge may have what little expertise exists. As one former official remarked about Somalia’s clan system, “We simply do not have enough of this kind of knowledge.” 76 Such bureaucratic ignorance is structural: the government does not have the resources to learn everything about everywhere, and expertise is gained by working on problems of interest to U.S. policy: when this shifts suddenly, expertise is often lacking. Here

73 Interview with former Obama administration official, December 15, 2013. 74 Kruzel, “More a Chasm Than a Gap,” 180. 75 Richard Haass, “The Irony of American Strategy,” Foreign Affairs , May/June 2013, https://www.foreignaffairs.com/articles/united-states/2013-04-03/irony-american-strategy 76 Kruzel, “More a Chasm Than a Gap,” 179.

28 area studies has value over international relations theory, as the former can fill gaps that policymakers immediately perceive. 77

P7. Responding to Policy Failure. Another window opens when a policy that seems set is clearly not working. Judging success or failure can be exceptionally difficult (the containment of

Iraq was widely deemed to be failing in the late 1990s, when in fact Saddam Hussein had secretly suspended his nuclear program) and politically contentious. In some instances, however, the inadequacy of a current approach is clear to all but the most hidebound. In Iraq by 2006, the collapse of order and the outbreak of all-out civil war made manifest U.S. policy failures despite adamant Bush administration denials: a “self-evident disaster,” as one official put it. 78 The administration and military began casting around for new approaches – ideas for a counterinsurgency-based surge advanced by a mix of academics, think tank officials, and military “outsiders” found a receptive audience. 79

Propositions on Injecting Academic Ideas into the Policy Process. Even if one conducts relevant and timely academic research, it will not influence policy if it does not come to the attention of policymakers. Some academics (and think tankers) often assume that publication equals influence, but in reality, publication of an academic article is often the beginning, not the end, of the process of influencing the debate. Policymakers are busy people – and the more senior ones are exceptionally busy. Indeed, perhaps the biggest difference between scholars, who can linger over problems, and policymakers is that the latter have little time to go deep on issues – a problem that grows as one advances up the policy food chain. The time they have to read a short memo is limited; their time for a longer paper almost non-existent. One official put it, “If I make

77 Avey and Desch, “What Do Policymakers Want from Us?” 78 Interview with former Obama and Clinton administration official, January 14, 2014, Washington, DC. 79 Kaplan, The Insurgents, for a review.

29 it through the Post , I’m happy.” 80 As one respondent to the Avey and Desch survey found, “the challenges are real time, in the moment, and situational.” 81 Scholars may not be comfortable with the steps required to land one’s arguments and findings on the desk of a busy policymaker, but it is necessary if one hopes to influence policy.

To make it to the next step, scholars must offer nuanced arguments, choose the proper outlet (or outlets), try to set the agenda with high-profile publications, influence the views of lower-level officials and intelligence officers, and develop personal relationships with policymakers.

P8. Offer Nuanced Arguments. When explaining the policy implications of their research, scholars will often simply prescribe a broad course of action consistent with their argument and findings, but this is not enough if scholars wish to influence ongoing debates. Policymakers need to know what specifically they should do differently on Monday morning and authors need to make the importance of their argument immediately obvious to busy policymakers. Scholars should seek to answer the following questions: What is the conventional wisdom surrounding a subject or the United States’ current approach? Does the author agree or disagree with it? If not, what should be done instead? Is the policy being advocated inconsistent with other measures? If not, is a both/and framing better than either/or? Does the authors’ research tell us everything we need to know about a policy question, or is it only one piece of a larger puzzle? Such nuanced analysis might appear less exciting than sweeping generalizations about how the country should deal with the world, but if the author does not do this work herself, then she is leaving too much work to the policymaker—work that will probably never be done.

80 Interview with former Obama and Clinton administration official, January 14, 2014, Washington, D.C. 81 Avey and Desch, “What Do Policymakers Want from Us?”

30 P9. Outlet Matters. Scholarship published in outlets read by policymakers will be more likely to have an influence. Shorter is always better. Even mid-level officials like an NSC Senior

Director working on Russia are more likely to read the New York Times, The Washington Post, or the Carnegie Endowment Moscow website rather than a long academic article. Scholars must also be attentive to the time required to publish in various venues in order to meet the timing requirements discussed above. It will be difficult to hit a fleeting target of opportunity if one chooses to publish exclusively in a slow-moving peer reviewed journal. This does not mean scholars should avoid scholarly journals if they want to be policy relevant. The advantage scholars bring to the policy debate is that their views are informed by in-depth, rigorous research.

An ideal approach would be to transform a book or a scholarly journal article into a shorter, more accessible piece that appears in Foreign Affairs . This, too, would be boiled down to one or several opinion pieces, blog entries, tweets, and so on.

P10. Set the Agenda. Publications in high-profile newspapers often compel policymaker attention. Former CIA director George Tenet recalled, “In both administrations that I worked for, what was in the news would often drive policy makers’ agenda. That was often the first thing they wanted to talk about.” 82 Scholars might be fortunate enough to publish directly in high- profile outlets or have their work read and promoted by established voices such as New York

Times columnists. Policymakers’ short-time horizons and impatience leads such works to gain more prominence than a sober, detailed in-depth study on the same issue.

Even when pieces do not grab the attention of principals, outside analysis often made its way into the process through desk officers or other lower-ranking officials whose work, in combination and with many interruptions, helps set the agenda for their bureaucratic and political superiors. Desk officers on Iran at the Defense Department, for example, read almost all the

82 Tenet, At the Center of the Storm, 30.

31 opinion pieces published on Iran in major newspapers, Foreign Affairs, and Foreign Policy, among other popular news sites. (Major academic journal articles, however, are too long). If they find persuasive arguments, this may find its way into read-aheads for senior leaders, be incorporated into options papers, and so on. Pieces that are interpretive, offering ways to understand a complex or unexpected phenomenon (e.g. the Arab spring) can be particularly useful for agenda setting, helping policymakers ask new questions or consider new approaches. 83

P11. Influence Intelligence. Intelligence analysis filters up and sometimes reaches the

President himself. 84 Policymakers may not read outside analysis regularly, but many intelligence analysts do. This contributes to their “all source” assessments and otherwise is part of the information they pass to policymakers. Intelligence often doesn’t matter when the great foreign policy decisions are made. And, as one official noted, even when it matters, this route “is a back door, and it gets pretty diluted.” 85 But at times intelligence influences these decisions, and it certainly has an impact on a host of lesser issues. 86 Intelligence analysts, moreover, ask questions similar to what scholars ask – why does Putin do what he does? – though they use different, and usually more fine-grained, information to answer. So by influencing intelligence analyses, academics can indirectly shape policy.

P12. Develop Personal Relationships with Policymakers and Intelligence Analysts.

Scholarship by people with personal relationships with policymakers and intelligence analysts is more likely to influence policy. It is human nature to pay attention to the activities of people one knows. One is more likely to read a book or an article written by a friend or former colleague than one written by a complete stranger. Scholars interested in conducting policy relevant work,

83 Interview with former Obama and Clinton administration official, January 14, 2014, Washington, D.C. 84 Interview with former Obama administration official, December 15, 2013, Washington, D.C. 85 Interview with former Obama and Clinton administration official, January 14, 2014, Washington, D.C. 86 For a sophisticated discussion of what is reasonable to expect of intelligence, see Betts, Enemies of Intelligence.

32 therefore, would be well advised to spend time in Washington DC and take other steps to develop personal relationships with policymakers. This will improve the chances that their future research makes its way into the system. If one follows the advice provided above, developing relationships with those on the inside should be easier as scholars conducting the most policy- relevant work will be more likely to be invited to meet with or brief government officials.

P13. Join Bureaucratic Warfare. Before the policy line is set, bureaucracies may quarrel over the correct approach – and this offers outsiders a chance to have their voices heard. Those on the inside may turn to outside work to bolster their case or, more rarely, inform their thinking.

One former official noted that outside work such as the public writings of key figures like

Kissinger and Brzezinski shaped the internal debate on NATO enlargement.87 Many within a bureaucracy may want to change policy but cannot use the traditional system to challenge the policy line. As such, they may look to outside analysis to reveal faults and propose alternatives: arguments they can then send up the chain because, ironically, they are outside the chain itself.

At times they may even quietly solicit such pieces (what one interviewee described as “calling in artillery on your own position”). 88

Outsiders are often used as stalking horses for controversial policies. For example, to push a change in strategy in Iraq, White House officials (including prominent academic Peter

Feaver who was on leave from his academic job) arranged briefings from think tankers and scholars who held similar views: it was far easier for the outsiders to openly recommend dramatic change than it was for those in government. 89 Government officials can advance – and then disavow – outside work, giving it more freedom to advocate controversial ideas. Similarly,

87 Interview with former Clinton and Bush administration official, Washington, DC January 14, 2014, Washington, D.C. 88 Interview with former State Department official, Washington, D.C., February 15, 2014. 89 Baker, Days of Fire.

33 via “Track II” or similar efforts outsiders can convene people (e.g. someone affiliated with a controversial organization like the Muslim Brotherhood) that the government itself cannot invite due to political controversy. 90

This recommendation calls, therefore, for academics to write for particular parts of the government rather than the foreign policy establishment at large. A good rule is to link one’s argument and variables to an office at the State or Defense Departments, the CIA, the NSC, or another such part of the bureaucracy that has an institutional interest in the issue being advocated. If the argument can’t be linked to any such office, it won’t be part of the bureaucratic debate.

This is not to argue that academics should produce scholarship with the goal of appealing to a particular bureaucracy. After all, as we stated above, independence and intellectual freedom are among policymakers’ greatest virtues. Rather, these steps should be reversed. Scholars should conduct their scholarly analysis and then, if they hope to influence policy, seek to identify an ally in the government who might have an interest in championing the findings.

Vital is recognizing the interests of particular bureaucracies and the data stream that feeds them. For those in the Defense world, “If it's not in the Early Bird, it doesn't matter.” 91 But even more important is recognizing what more specialized communities consume daily. For example, the blog Lawfare, which writes for the legal national security community, is read by far fewer readers than, say, ForeignPolicy.com, but its readers include the most senior officials from this world. Similarly, many officials working on arms control and nuclear issues consult Arms

Control Wonk. These blogs focus on specialized subjects with technical information and can,

90 Singer, “Factories to Call Our Own.” 91 The National Journal assumed the “early bird” role in 2013. See http://www.nationaljournal.com/early-bird.

34 therefore, be more useful to officials than outlets aimed at reaching a broader audience. 92 The good news is that there are many marketplaces of ideas: the bad news is that policymakers can choose the ones they frequent and ignore others.

Some readers may find the notion of joining bureaucratic warfare distasteful or too inside-the-beltway, but as we will next show we have reasons to believe that this is an effective way to get one’s research into the system and that there are ways to guard against the possible downside risks of this approach.

V. A Case Study: Nuclear Posture

An illustrative case study demonstrates many of the propositions we identify above. One of us (Kroenig) has recently conducted research showing that countries with nuclear superiority, i.e., more nuclear weapons than their opponents, are more likely to achieve their goals in international crises. This research was first presented at the American Political Science

Association Annual Meeting in the fall of 2009 and was published in International Organization in January 2013. 93 The research has a relevant dependent variable – there are many offices spread throughout the U.S. government focused on nuclear deterrence, including the Defense

Policy and Strategy office of the NSC, the Nuclear and Missile Defense office in the Defense

Department, U.S. Strategic Command, the Strategic Deterrence and Nuclear Integration office in the U.S. Air Force, Strategic Plans and Policy in the Joint Staff, and the National Nuclear

Security Administration in the Department of Energy. The research also had an independent

92 Interview with Department of Energy official, Washington, D.C., January 22, 2014. 93 Matthew Kroenig, “Nuclear Superiority or the Balance of Resolve? Explaining Nuclear Crisis Outcomes,” Annual Meeting of the American Political Science Association, Toronto, Canada, 2009; Matthew Kroenig, “Nuclear Superiority and the Balance of Resolve: Explaining Nuclear Crisis Outcomes,” International Organization 67, no. 1 (January 2013): 141-71.

35 variable that policymakers can manipulate: the size of the U.S. nuclear arsenal. While the academic study included methods in which most policymakers would not be conversant, including game theory and statistical analysis, the author also gave public talks in Washington

DC and published a number of related opeds, policy journal articles, and blogs reporting on the findings of the research. 94

The research was also conducted at a propitious time. Shortly after taking office in Spring

2009, U.S. President Barack Obama made the creation “of a world without nuclear weapons” an explicit goal of U.S. policy. To pursue this goal, the administration had planned to pursue arms control agreements with Russia to reduce the size of U.S. and Russian arsenals and to codify

America’s reduced reliance on nuclear weapons in the Congressionally-mandated Nuclear

Posture Review (NPR), scheduled for release in the Spring of 2010.

While these objectives had strong political support among the President’s closest advisers, and in the parts of the U.S. government responsible for nonproliferation and arms control, others were more skeptical. Many offices in the Department of Defense and the White

House’s own NSC, responsible for defense and deterrence, and many Republicans on Capitol

Hill believed that further U.S. nuclear reductions were unwise. Some believed the arsenal had already reached a bare minimum and that further reductions would harm America’s nuclear deterrent. Others were skeptical that agreements with Moscow were helpful or could be trusted.

Still others assessed that cuts could be acceptable only if the administration vowed to invest

94 Matthew Kroenig, “Nuclear Zero? Why Not Nuclear Infinity? The World Would Not Be Safer if the United States Had No Nuclear Weapons,” The Wall Street Journal , July 30, 2011, http://www.wsj.com/articles/SB10001424053111903554904576464571545999158; Matthew Kroenig, “The Case for Overkill: Obama’s Wrong, America Does Need Thousands of Nukes,” Foreign Policy, June 19, 2013, http://foreignpolicy.com/2013/06/19/the-case-for-overkill/; Matthew Kroenig, “Think Again: American Nuclear Disarmament,” Foreign Policy , September 3, 2013, http://foreignpolicy.com/2013/09/03/think-again-american- nuclear-disarmament/; Matthew Kroenig, Miriam Krieger, and Hans Noel, “Why Nuclear Superiority Matters for Compellence,” The Monkey Cage , December 3, 2014, http://www.washingtonpost.com/blogs/monkey- cage/wp/2014/12/03/why-nuclear-superiority-matters-for-compellence/

36 significantly in America’s nuclear infrastructure to ensure the continued existence of a robust nuclear posture going forward.

When given the opportunity, Kroenig joined bureaucratic warfare to advance his research. In the winter of 2010, Kroenig presented his research at an academic conference held at the University of California’s center in Washington DC. A congressional staffer was in the audience, was taken by the research, and helped to set up a briefing to Republican staff on the

Hill, many of whom had bosses with an interest in killing, or at least extracting more concessions from the Obama administration over, the New START arms control negotiations, which were then well underway with Russia.

At about the same time, Kroenig was contacted by a former colleague, then at the NSC, who was working on the NPR and asked Kroenig if he could provide examples of some of the best, recent academic work on nuclear deterrence. Kroenig sent his contact several articles, including research by Keir Lieber and Daryl Press among others, as well as his APSA conference paper. 95 Unlike some other offices just down the hall in the Eisenhower Executive Office

Building (EEOB), this office was somewhat skeptical about the need for further nuclear reductions. Kroenig was later told that the paper was included in read-aheads for interagency meetings on the NPR and was even closely read and marked up by President Obama himself.

This last claim is hard to fathom considering the many demands on a President’s time. Given that

Obama is a cerebral president with a life-long interest in nuclear issues and the trustworthiness of the source, however, the claim is at least plausible.

In the end, the paper did not have a decisive bearing on the policy outcome. On April 8,

2010, the Obama administration signed the New START agreement with Russia, promising to

95 Keir A. Lieber and Daryl Press, “The Nukes We Need: Preserving the American Nuclear Deterrent,” Foreign Affairs 88, no. 6 (November/December 2009): 39-51.

37 reduce the size of the U.S. nuclear arsenal to 1,550 warheads before 2018, and the treaty was approved by the Senate. Three days later, the NPR was released, which vowed to reduce

America’s reliance on nuclear weapons.

The debate did not end in the spring of 2010, however. Before the ink was dry on the

NPR, the administration announced a new 90-day study to look into the possibility of further

U.S. nuclear reductions. As this debate played out, Kroenig’s research on nuclear superiority was once again in demand. Over the next several years, Kroenig was invited to give briefings on the subject to the NSC, the Pentagon, U.S. Strategic Command, Lawrence Livermore National

Laboratory, and on Capitol Hill. In March 2011, Senator Jon Kyl from Arizona cited the research, saying “It follows, too, that an effective nuclear arsenal discourages challenges to the

United States and its interests… Georgetown Professor Matthew Kroenig concludes, and I quote,

‘that American nuclear superiority has helped to deter adversaries from initiating militarized disputes against the United States in the past.’” 96 In the end, the prospect for additional U.S. nuclear reductions was halted more than anything else by worsening relations with Russia and the impossibility of further arms control negotiations, but this is an academic and policy debate that is likely to be with us for years to come.

In sum, this research met our standard for policy relevance; it was part of the discussion even if it did not have a decisive bearing on the policy outcome (at least not yet). It was designed with a relevant dependent variable and a manipulable independent variable and even though it employed sophisticated research methods, the findings were made accessible through public presentations, policy articles, opeds, and blogs. The research had the good fortune to be

96 Kyl, Senator Jon. “2011 Carnegie International Nuclear Policy Conference.” Lecture, from the Carnegie Endowment for International Peace, Washington, D.C., March 29, 2011, http://carnegieendowment.org/files/Senator_Jon_Kyl.pdf.

38 published at a time when information was needed on the subject and the author took the necessary steps to get it into the policy debate, including drawing on personal relationships and joining bureaucratic warfare.

VI. The Downsides of Policy Relevance

There are, of course, potential downsides to policy relevant research that scholars must guard against, but we believe that these problems are not insurmountable. In particular, each of the four large approaches that we recommend for achieving scholarly relevance carries with it potential dangers.

First, in an attempt to design policy relevant research, the quality of academic research as a whole may suffer as individual scholars, or the field as a whole, systematically avoid less policy relevant, but important dependent variables, like international stability; less policy relevant, but potentially important structural independent variables, like the international balance of power; and rigorous, but hard to understand, social science research methods.

We believe, however, that, on reflection, each of these risks is actually quite low. First, for the field as a whole, diverse scholarly interests and professional incentives will continue to incentivize many scholars to conduct research in less policy relevant domains. Moreover, at the individual level, peer review will demand that research meets the required standards and if scholars fail to consider important causal variables or employ appropriate research methods, the research will not see the light of day. Finally, as the work of Avey and Desch among other suggests, the field has moved away from policy relevance and a shift back toward it would still mean that the vast bulk of scholars were not directly trying to influence the policy debate.

39 Second, scholars may be tempted to produce substandard scholarship as they rush to publication in order to influence ongoing events. Once again, however, we believe this risk is less severe than it initially may appear. Once again, peer review provides a protection against shoddy scholarship making it into top scholarly journals. Alternatively, a scholar may choose to publish insights and analysis in a nonpeer-reviewed policy journal or oped before having all the necessary information, but rather than a problem, this route should be encouraged, particularly if the work is a spinoff of or builds on a broader foundation of peer-reviewed scholarship. The U.S. government often has to make time-sensitive decisions with or without all the necessary information and those deliberations are almost certainly improved by having highly-trained and expert scholars contribute their best insights to the marketplace of ideas, even if these latest insights have not been fully tested by academic standards.

Third, in a bid to establish personal relationships, or join bureaucratic warfare, a scholar may rig his or her analysis in order to produce an intended result. As Janice Stein contends, “We will be seduced by the proximity to power and shade what we may say in order to retain access.” 97 We believe this may be the most serious danger listed here, but, again, the situation is not hopeless. To begin, this problem is endemic in scientific research and is not unique to policy- relevant international relations research. Scholars often have incentives to produce specific conclusions in order to make a new contribution to an academic debate or to maintain one’s reputation and relationships in the field. As with other scholarly research, this danger will be kept in check by the peer-review process and by the scientific method as subsequent scholars critique and refine poor scholarship that was produced with the goal of producing a specific conclusion. In general, we urge scholars to do the peer-reviewed and foundational work first,

97 Stephen D. Krasner, Joseph S. Nye Jr., Janice Gross Stein, and Robert O. Keohane, “Autobiographical reflections on bridging the policy-academy divide,” Cambridge Review of International Affairs 22, no. 1 (March 2009): 111-28. For the quote from Stein, see p. 122.

40 establishing their own views before they enter the debate rather than letting the debate shape them.

Fourth, entering the policy debate may carry a price in productivity – at least as defined in traditional terms. By writing op-eds and blogs, by briefing policymakers, and by otherwise not working on peer-reviewed research, scholars will not have as much time to devote to more traditional academic work. On the other hand, however, their productivity may benefit in absolute terms and engagement with the policy world may generate questions and provide insights that directly lead to higher-quality academic publications in the future.

In sum, policy relevant research has downsides, but we believe that they can be mitigated by exacting disciplinary standards and the scientific process. Some scholars will feel more comfortable navigating the personal, professional, and, in some cases, moral dilemmas that come with producing and disseminating ideas that influence real-world events than others. This variance is desirable. Having scholars continue to do research for its own sake is important to the field as a whole. Nevertheless, we believe that scholars who are so inclined can successfully write for policy audiences while simultaneously maintaining the highest academic standards.

VII. Conclusion

This paper has sought to identify the characteristics of academic work that make it more or less policy relevant. Scholars interested in influencing policy debates should: design research that might produce actionable findings and recommendations, identify moments of ripeness in policy debates, and inject their ideas into the policy process. There are also steps the discipline

41 writ large could take to facilitate scholars on this quest. Ph.D. programs could offer training not just on how to write long scholarly articles, but also on how to write for different venues, including policy journals and opinion pieces. Academic departments could incorporate policy influence, in addition to scholarly influence, into tenure and promotion decisions.

For those who want to conduct rigorous academic scholarship and influence real-world debates, we provide the strongest possible encouragement and hope our advice proves helpful.

This is not to say that we have provided a formula for reliably influencing important policy debates. Scholars seeking to be policy relevant should prepare for (repeated) failure. We have written pieces that we hope are jargon free, appeared in widely-read publications, and engaged variables that policy makers control in a timely way. And though we were often ignored, we believe it is worth trying again. For at times scholars can nudge the ship of state, and in so doing they can make the world a better place.

42